Development and Application of CM-GATS Algorithms in Solving Large Multilevel Hierarchical Network Design Problems

This study discusses a method that overcomes the size limitations in solving large size multilevel hierarchical hub location problem with a single allocation by using a stepwise solving approach called CM-GATS. This model uses a clustering algorithm to divide a given network into several small clusters with a manageable number of nodes then separately use hybrid heuristic algorithm based on Genetic Algorithm and Tabu Search to solve each cluster. The applicability of the proposed model in solving large multilevel hierarchical network design problems was verified by providing calculation results of CM-GATS in a large network containing up to 2,000 nodes.